SEARCH · Engineering Papers

Results for “variability prediction”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 505 records · Page 28

All sources NASA NTRS DOE OSTI

DOE OSTI2021

Explore Spatio‐Temporal Learning of Large Sample Hydrology Using Graph Neural Networks

Abstract Streamflow forecasting over gauged and ungauged basins play a vital role in water resources planning, especially under the changing climate. Increased availability of large sample hydrology data sets, together with recent advances in deep learning techniques, has presented new opportunities to explore temporal and spatial patterns in hydrological signatures for improving streamflow forecasting. The purpose of this study is to adapt and benchmark several state‐of‐the‐art graph neural network (GNN) architectures, including ChebNet, Graph Convolutional Network (GCN), and GraphWaveNet, for end‐to‐end graph learning. We explicitly represent river basins as nodes in a graph, learn the spatiotemporal nodal dependencies, and then use the learned relations to predict streamflow simultaneously across all nodes in the graph. The efficacy of the developed GNN models is investigated using the Catchment Attributes and MEteorology for Large‐sample Studies (CAMELS) data set under two settings, fixed graph topology (transductive learning), and variable graph topology (inductive learning), with the latter applicable to prediction in ungauged basins (PUB). Results indicate that GNNs are generally robust and computationally efficient, achieving similar or better performance than a baseline model trained using the long short‐term memory (LSTM) network. Further analyses are conducted to interpret the graph learning process at the edge and node levels and to investigate the effect of different model configurations. We conclude that graph learning constitutes a viable machine learning‐based method for aggregating spatiotemporal information from a multitude of sources for streamflow forecasting

Sun, Alexander Y.↗

DOE OSTI2020

The role of chemical potential in compensation control in Si:AlGaN

Reduction in compensation in Si-doped Al-rich AlGaN is demonstrated via chemical potential control (CPC). The chemical potentials and the resulting formation energies of carbon on the nitrogen site (C N ) and cation vacancy complex with Si (V III + nSi III ) were related to growth variables through a thermodynamic supersaturation model, which quantitatively predicted the incorporation of C N and the generation of the V III + nSi III complex. The compensation “knee” behavior, i.e., decreasing conductivity with increasing Si incorporation beyond a certain concentration, was successfully controlled. The maximum free carrier concentration was improved by impeding the formation of V III + nSIi III complexes under III-richer conditions, while the impurity compensation by C N was reduced by making the growth environment N-richer. The results of Hall effect measurement and photoluminescence agreed well with quantitative theoretical predictions of the CPC model. Based on the developed model, the highest conductivity of 160 Ω -1 cm -1 with free carrier concentration of 3 × 10 19 cm -3 in Al 0.7 Ga 0.3 N ever reported was achieved on single crystal AlN substrates. The demonstrated predictive power of the CPC model should greatly reduce the empirical analysis or iterative experimentation that would otherwise be necessary.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2025

Nanometer-thick ultraflat cantilever resonators

Nanomechanical devices made from ultrathin materials are transforming diverse fields, including sensing, signal processing, and quantum technologies. However, as these materials become thinner, their low bending rigidity poses significant fabrication challenges, and achieving nanometer-thick flat cantilevers with consistent and predictable mechanical responses has remained elusive despite decades of research. Here we present nanometer-thick, ultraflat cantilever resonators fabricated using atomic layer deposition. By effectively mitigating the effects of uncontrollable built-in strain and geometric disorder, the ultraflat nanocantilevers exhibit resonance frequencies closely aligned with thin-plate theory predictions and display low sample-to-sample variability. These cantilevers maintain mechanical stability in both vacuum and air environments, even at large length-to-thickness ratios of up to 3000. The ultraflat nanocantilevers are approaching the thickness limit, beyond which thermal fluctuations at room temperature can spontaneously induce random ripples in otherwise flat films.

nanofabrication↗

DOE OSTI2021

Extracting Resilience Metrics From Distribution Utility Data Using Outage and Restore Process Statistics

Resilience curves track the accumulation and restoration of outages during an event on an electric distribution grid. We show that a resilience curve generated from utility data can always be decomposed into an outage process and a restore process and that these processes generally overlap in time. We use many events in real utility data to characterize the statistics of these processes, and derive formulas based on these statistics for resilience metrics such as restore duration, customer hours not served, and outage and restore rates. The formulas express the mean value of these metrics as a function of the number of outages in the event. We also give a formula for the variability of restore duration, which allows us to predict a maximum restore duration with 95% confidence. Overall, we give a simple and general way to decompose resilience curves into outage and restore processes and then show how to use these processes to extract resilience metrics from standard distribution system data.

24 POWER TRANSMISSION AND DISTRIBUTION↗

DOE OSTI2023

Chaconne: A Statistical Approach to Nonlocal Compression for Supervised Learning, Semi-Supervised Learning, and Anomaly Detection

This project developed a novel statistical understanding of compression analytics (CA), which has challenged and clarified some core assumptions about CA, and enabled the development of novel techniques that address vital challenges of national security. Specifically, this project has yielded the development of novel capabilities including 1. Principled metrics for model selection in CA, 2. Techniques for deriving/applying optimal classification rules and decision theory to supervised CA, including how to properly handle class imbalance and differing costs of misclassification, 3. Two techniques for handling nonlocal information in CA, 4. A novel technique for unsupervised CA that is agnostic with regard to the underlying compression algorithm, 5. A framework for semisupervised CA when a small number of labels are known in an otherwise large unlabeled dataset. 6. The academic alliance component of this project has focused on the development of a novel exemplar-based Bayesian technique for estimating variable length Markov models (closely related to PPM [prediction by partial matching] compression techniques). We have developed examples illustrating the application of our work to text, video, genetic sequences, and unstructured cybersecurity log files.

99 GENERAL AND MISCELLANEOUS↗

DOE OSTI2021

Spatially-Explicit Prediction of Capacity Density Advances Geographic Characterization of Wind Power Technical Potential

Mounting interest in ambitious clean energy goals is exposing critical gaps in our understanding of onshore wind power potential. Conventional approaches to evaluating wind power technical potential at the national scale rely on coarse geographic representations of land area requirements for wind power. These methods overlook sizable spatial variation in real-world capacity densities (i.e., nameplate power capacity per unit area) and assume that potential installation densities are uniform across space. Here, we propose a data-driven approach to overcome persistent challenges in characterizing localized deployment potentials over broad extents. We use machine learning to develop predictive relationships between observed capacity densities and geospatial variables. The model is validated against a comprehensive data set of United States (U.S.) wind facilities and subjected to interrogation techniques to reveal that key explanatory features behind geographic variation of capacity density are related to wind resource as well as urban accessibility and forest cover. We demonstrate application of the model by producing a high-resolution (2 km × 2 km) national map of capacity density for use in technical potential assessments for the United States. Our findings illustrate that this methodology offers meaningful improvements in the characterization of spatial aspects of technical potential, which are increasingly critical to draw reliable and actionable planning and research insights from renewable energy scenarios.

17 WIND ENERGY↗

DOE OSTI2022

Role of Sea‐Surface Salinity in Simulating Historical Decadal Variations of Atlantic Meridional Overturning Circulation in a Coupled Climate Model

Abstract Extending climate prediction from seasonal to decadal timescales requires realistic initialization of not only upper ocean heat content but also the Atlantic meridional overturning circulation (AMOC). However, it remains a major challenge to realistically initialize AMOC in a coupled system while also maintaining a balanced atmosphere‐ocean initial state. This study demonstrates the feasibility of generating fully coupled historical states with realistic AMOC variability. Employing a forced ocean—sea‐ice (FOSI) model simulation as the “truth,” we show reproducibility of key features of historical AMOC decadal variability in a fully coupled model by restoring sea‐surface salinity, in addition to sea‐surface temperature restoring widely used in seasonal prediction. The atmospheric state of the restored coupled model solution is much closer to that of the free coupled simulation than to the observations used in FOSI, pointing to potential advantages of using this approach for initializing decadal predictions with reduced ocean initialization shock.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2024

Bayesian inference of structured latent spaces from neural population activity with the orthogonal stochastic linear mixing model

The brain produces diverse functions, from perceiving sounds to producing arm reaches, through the collective activity of populations of many neurons. Determining if and how the features of these exogenous variables (e.g., sound frequency, reach angle) are reflected in population neural activity is important for understanding how the brain operates. Often, high-dimensional neural population activity is confined to low-dimensional latent spaces. However, many current methods fail to extract latent spaces that are clearly structured by exogenous variables. This has contributed to a debate about whether or not brains should be thought of as dynamical systems or representational systems. Here, we developed a new latent process Bayesian regression framework, the orthogonal stochastic linear mixing model (OSLMM) which introduces an orthogonality constraint amongst time-varying mixture coefficients, and provide Markov chain Monte Carlo inference procedures. We demonstrate superior performance of OSLMM on latent trajectory recovery in synthetic experiments and show superior computational efficiency and prediction performance on several real-world benchmark data sets. We primarily focus on demonstrating the utility of OSLMM in two neural data sets: μ ECoG recordings from rat auditory cortex during presentation of pure tones and multi-single unit recordings form monkey motor cortex during complex arm reaching. We show that OSLMM achieves superior or comparable predictive accuracy of neural data and decoding of external variables (e.g., reach velocity). Most importantly, in both experimental contexts, we demonstrate that OSLMM latent trajectories directly reflect features of the sounds and reaches, demonstrating that neural dynamics are structured by neural representations. Together, these results demonstrate that OSLMM will be useful for the analysis of diverse, large-scale biological time-series datasets.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2020

Seasonal mixing from intermittent flow drives concentration-discharge behaviour in a stream affected by coal mine drainage

Abandoned mining operations continue to severely degrade many ecosystems worldwide by releasing acidic water and/or heavy metals into surface and groundwater. Contaminant concentrations in affected streams vary with discharge in patterns that reflect both geochemical reactions and variable mixing of contaminated and non-contaminated waters. However, controls on concentration-discharge (C-Q) patterns remain unclear, particularly for constituents that experience changing solubility across redox and pH gradients. Understanding the C-Q behaviour of contaminants aids in predicting both downstream transport and effects on aquatic life under variable flow. Here, we examined the C-Q behaviours of non-reactive (Na, K, Ca, Mg, Cl - ) and reactive (Fe, Mn, Al, H + , SO 4 2- ) solutes in a stream contaminated with acid mine drainage in northeastern Ohio, USA. Concentration-discharge patterns at the watershed outlet primarily reflected mixing of contaminated baseflow with intermittent inputs of high pH water draining from a passive limestone treatment system into the stream. The treatment system acted as an ephemeral tributary that mitigated contamination in the stream by diluting solutes, raising pH, and driving metal precipitation, but only when flow was present during wet seasons. Consequently, AMD-derived reactive solutes (H + , Fe, Mn, Al) decreased with increasing stream discharge while relatively conservative solutes (e.g., Ca, Mg, K, Na) decreased only slightly or were chemostatic. This study highlights both the unique C-Q patterns of reactive solutes when compared to those of non-reactive solutes and the potential for intermittent streams to control C-Q behaviour in headwater catchments.

58 GEOSCIENCES↗

DOE OSTI2022

Analytics-at-scale of Sensor Data for Digital Monitoring in Nuclear Plants (3 rd Annual Report)

Nuclear power plants collect and store large volumes of heterogeneous data from various components and systems. With recent advances in machine learning (ML) techniques, these data can be leveraged to develop diagnostic and short-term forecasting models to better predict future equipment condition. Maintenance operations can then be planned in advance whenever degraded performance is predicted, thus resulting in fewer unplanned outages and the optimization of maintenance activities. This enables lower maintenance costs and improves the overall economics of nuclear power. This report primarily focuses on developing a short-term forecasting process that leverages a feature selection process to distill large volumes of heterogeneous data and predict specific equipment parameters. A variety of feature selection methods, including Shapley Additive Explanations (SHAP) and variance inflation factor (VIF), were used to select the optimal features as inputs for three ML methods: long short-term memory (LSTM) networks, support vector regression (SVR), and random forest (RF). Each combination of model and input features was used to predict a pump bearing temperature both 1 and 24 hours in advance, based on actual plant system data. The optimal inputs for the LSTM and SVR were selected using the SHAP values, while the optimal input for the RF consisted solely of the response variable itself. Each model produced similar 1-hour-ahead predictions, with root mean square errors (RMSEs) of roughly 0.006. For the 24-hour-ahead predictions, differences could be seen between LSTM, SVR, and RF, as reflected by model performances of 0.036 ± 0.014, 0.0026 ± 0, and 0.063 ± 0.004 RMSE, respectively. As big data and continuous online monitoring become more widely available, the proposed feature selection process can be used for many applications beyond the prediction of process parameters within nuclear infrastructure. This report summarizes the Fiscal Year 2021 research progress encompassing the (1) data cleaning and feature selection necessary for ML applications; (2) development of short-term forecasting models to predict future plant process parameters for both single and multiple time steps ahead; and (3) validation of the feature selection methods and short-term forecasting models given new data from different systems.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2022

Development of Short-Term Forecasting Models Using Plant Asset Data and Feature Selection

Nuclear power plants collect and store large volumes of heterogeneous data from various components and systems. With recent advances in machine learning (ML) techniques, these data can be leveraged to develop diagnostic and short-term forecasting models to better predict future equipment condition. Maintenance operations can then be planned in advance whenever degraded performance is predicted, thus resulting in fewer unplanned outages and the optimization of maintenance activities. This enables lower maintenance costs and improves the overall economics of nuclear power. This paper focuses on developing a short-term forecasting process that leverages a feature selection process to distill large volumes of heterogeneous data and predict specific equipment parameters. A variety of feature selection methods, including Shapley Additive Explanations (SHAP) and variance inflation factor (VIF), were used to select the optimal features as inputs for three ML methods: long short-term memory (LSTM) networks, support vector regression (SVR), and random forest (RF). Each combination of model and input features was used to predict a pump bearing temperature both 1 and 24 hours in advance, based on actual plant system data. The optimal inputs for the LSTM and SVR were selected using the SHAP values, while the optimal input for the RF consisted solely of the response variable itself. Each model produced similar 1-hour-ahead predictions, with root mean square errors (RMSEs) of roughly 0.006. For the 24-hour-ahead predictions, differences could be seen between LSTM, SVR, and RF, as reflected by model performances of 0.036 +- 0.014, 0.0026 +- 0, and 0.063 +- 0.004 RMSE, respectively. As big data and continuous online monitoring become more widely available, the proposed feature selection process can be used for many applications beyond the prediction of process parameters within nuclear infrastructure.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2021

Multi-fidelity thermal modeling of laser powder bed additive manufacturing

Laser powder bed fusion (LPBF) Additive manufacturing (AM) has attracted interest as an agile method of building production metal parts to reduce design-build-test cycle times for systems. However, predicting part performance is difficult due to inherent process variabilities. This makes qualification challenging. Computational process models have attempted to address some of these challenges, including mesoscale, full physics models and reduced fidelity conduction models. The goal of this work is credible multi-fidelity modeling of the LPBF process by investigating methods for estimating the error between models of two different fidelities. Two methods of error estimation are investigated, adjoint-based error estimation and Bayesian calibration. Adjoint-based error estimation is found to effectively bounding the error between the two models, but with very conservative bounds, making predictions highly uncertain. Bayesian parameter calibration applied to conduction model heat source parameters is found to effectively bound the observed error between the models for melt pool morphology quantities of interest. However, the calibrations do not effectively bound the error in heat distribution.

36 MATERIALS SCIENCE↗

DOE OSTI2023

Swelling and permeability effects during propellant cookoff

Large rocket motors may violently explode when exposed to accidental fires. Even hot metal fragments from a nearby accident may penetrate the propellant and ultimately cause thermal ignition. A mechanistic understanding of heated propellants leading to thermal runaway is a major unsolved problem. Here we show that thermal ignition in propellants can be predicted using a universal cookoff model coupled to a micromechanics pressurization model. Our model predicts the time to thermal ignition in cookoff experiments with variable headspace volumes. Furthermore, we found that experiments with headspace volumes are more prone to deformation which distorts pores and causes increased permeability when the propellant expands into this headspace. Delayed ignition with larger headspace volume correlates with lower headspace pressures during decomposition. We found that our predictions matched experimental measurements best when the initial propellant was impermeable to gas flow rather than being permeable. Similar behavior is expected with other energetic materials with rubbery binders. Our model is validated using data from a separate laboratory. We also present an uncertainty analysis using Latin Hypercube Sampling (LHS) of thermal ignition caused by a steel fragment embedded in the propellant.

33 ADVANCED PROPULSION SYSTEMS↗

DOE OSTI2025

Analyzing the Effects of the Interference between One- and Two-Body Currents on Neutrino-Nucleus Scattering in ACHILLES

Modeling improvements of neutrino–nucleus scattering are an essential aspect of reducing systematic uncertainties for current and future neutrino oscillation experiments. In particular, retaining quantum mechanical effects such as the interference between one-body (1p1h) and two-body (2p2h) currents in our cross-section models is an essential consideration. This work investigates the impact of this interference term on neutrino–argon scattering using the ACHILLES event generator, which directly implements the interference model developed by Lovato, Rocco, and Steinberg. Using fluxes relevant to SBND, the DUNE near detector, and MiniBooNE, charged- and neutral-current event samples are generated and predicted event rates are plotted against a variety of kinematic variables. For events simulated using the SBND flux, it is found that interference contributes approximately 10 \% to the total charged-current event rate, and serves to enhance the quasi-elastic peak. When considering events generated using the DUNE flux, resonance interactions contribute more to the predicted event rate, as the neutrino flux extends out to higher energies. The relative interference contribution subsequently decreases to roughly 7 \%. Neutral-current simulations show a higher rate of neutron knockout. Finally, a comparison of argon-to-carbon cross-section ratios is performed utilizing fluxes from SBND and MiniBooNE.

Pando, Nathan [Fermilab]↗

DOE OSTI2025

Analyzing the Effects of the Interference between One- and Two-Body Currents on Neutrino-Nucleus Scattering in ACHILLES

Pando, Nathan [Fermilab]↗

DOE OSTI2022

A Combined Machine Learning and High-Energy X-ray Diffraction Approach to Understanding Liquid and Amorphous Metal Oxides

Determining the structure-property relations of liquid and amorphous metal oxides is challenging, due to their variable short-range order and polyhedral connectivity. To predict chemically realistic structures, we have developed a Machine Learned, Gaussian Approximation Potential (GAP) for HfO 2 , with a focus on enhanced sampling of the training database and accurate density functional theory calculations. By using training datasets for the GAP model at the level of Density Functional Theory-Strongly Constrained and Appropriately Normed (DFT-SCAN) level of theory, our results show that the topology of both the low viscosity liquid and the amorphous form are dominated by edge-shared chains and small corner-shared rings of polyhedra. This topology is shown to be consistent with the structure of other liquid and amorphous transition metal oxides of variable ion size, such as TiO 2 and ZrO 2 . Current limitations of the ML-GAP modeling method for obtaining glass structures and future perspectives are also discussed.

36 MATERIALS SCIENCE↗

DOE OSTI2021

The Leading Modes of Asian Summer Monsoon Variability as Pulses of Atmospheric Energy Flow

Monsoon rainfall anomalies are often organized into dynamical modes that are more predictable than rainfall at specific locations. Here a linear response function (LRF) is constructed to extract the neutral modes (NM) of the dynamical atmospheric system and their optimal forcing that govern precipitation variability, providing a dynamical underpinning for the empirically derived, recurrent rainfall patterns. The leading mode (NM1) features a shift in the Atlantic ITCZ, a drying over the Indian Ocean and Indian subcontinent, and a “South-Flood-North Drought (SFND)” pattern over eastern China, a pattern inferable from changes in the atmospheric energy flow forced by the corresponding optimal forcing and further enhanced by the water vapor and cloud feedbacks. Moreover, the slow evolving component of the NM1 in response to CO2 forcing can be skillfully predicted by running the ocean heat uptake through the LRF, implicating a deterministic ocean dynamic source for the NM1-like trend under climate warming.

monsoon mode of variability, EBM, climate change, ↗

DOE OSTI2022

Training data selection for event classification in a highly variable environment

A problem of interest for nuclear nonproliferation is monitoring activities at nuclear facilities, where proliferation events may only take place a few times and often under variable conditions. Machine learning has revolutionized data analytics by enabling the use of measurable signatures to generate predictive models of facility operations. However, traditional methods for training these models require large, reliable data sets with labeled observations, a challenge for nonproliferation. Highly variable conditions further complicate this as events from training data may have occurred in conditions quite different from the event of interest. Our hypothesis is that when events occur in a highly variable environment, careful training data selection for each test event could outperform the standard approach of using all available training data. We developed a method to optimize training data selection for the given test event and applied it to predicting the power level of the High Flux Isotope Reactor (HFIR) at Oak Ridge National Laboratory. In this study, the reactor startup exhibits variability between occurrences due to natural variability in environmental conditions and operational procedures. Using a combination of analysis techniques, a similitude assessment was performed on data collected from HFIR to isolate clusters that were optimal for training a predictive model. Concepts such as dynamic time warping and Jaccard similarity were used in conjunction with clustering analysis. In order to validate this approach, the model was trained on every combination of unique training events and the predictive performance was compared to the performance using a subset of the training data selected by isolated clusters found through the similitude assessment.

Iyer, A↗